Untitled Document

 

Not a member yet? Register for full benefits!

Username
Password
Site Shop > The Emotion Machine: Commonsense Thinking, Artificial Intelligence, and the Future of the Human Mind




The author of this book, Marvin Minsky, is a professor of Computer Science at MIT. He helped found the field of AI itself, and is a staunch believer in strong AI systems ? artificial minds ? rather than the weak AI systems of pathfinding, expert systems and machine vision.

This, his latest book, seems a little out there at first, Emotions and humanity, discussed from an artificial intelligence master?

Yet, when examined more closely, you realise it is not odd at all. Minsky?s central premise, which runs right the way through this book, is clear, and does not waver. Contrary to popular conception, emotions aren't distinct from rational thought; rather, they are simply another way of thinking, one that computers could perform.

Throughout the book, it pulls steadfastly away from the conventional wisdom that the mind is a basically rational process distorted by emotion, or made more exciting by emotion, and instead embraces emotional states as just a different way to think, an expression of the mind no less valid than rational thought.

The Emotion Machine aims to find "more complex ways to depict mental events that seem simple at first." One of the key tenets explored is the supposition that consciousness remains unexplained because it is "one of those suitcase-like words that we use for many types of processes, and for different kinds of purposes."

Since consciousness is not a unity but involves separate mental components, "there is little to gain from wondering what consciousness 'is' - because that word includes too much for us to deal with all at once."

Throughout the book, it draws parallels between the human mind, and a computer program. One of the fundamental problems portrayed is the existence of weak AI systems today ? chess champions, othello champions, medical illness symptom diagnostics. Each can compete with, even outshine the best humans in this one chosen field, but try to step outside it, and they are lost. A human can play chess reasonably well, can drive a car, read books, cook up a feast, and leave a clean home each day. No current AI can do all this. They can only think in one domain at a time.

The book does make one curious diversion that is hard to explain. At no point does it look at the physical layout of the brain ? the neurons, the synapses, the neurotransmitters, the guts of the wiring and interconnections. Minsky?s reasons for this are that he feels he cannot focus on these components of a working mind as "research on this is advancing so quickly that any conclusion one might make today could be outdated in just a few weeks."

This results in an AI book which perhaps waxes more philosophical than direct physical comparisons, and some might say, this causes it to lack grounding for its flights of fancy. This does not seem to be the case. The arguments proposed are good, as far as they go. They open the mind, and leave it open, awaiting another to pick up the baton, dot the t?s cross the i?s, and take the work one step further.

At any rate, this book, if you have any interest in AI as in NPCs and agents who truly can think, then this is a very good starting point, and maybe leaping pad for your own theories and ideas.

 

Member Reviews

Reviews by our members. Become a member today, and submit a review!